Johannesburg, 13 Jun 2018
A few weeks after Google's annual developer conference, most of us have seen the fascinating yet frightening videos of Google Assistant phoning a restaurant to make a reservation.
Mike Geyser, BBD R&D lead, Google Developers Expert in Web technologies and all-round JavaScript enthusiast, attended Google I/O 2018 and shared his mind-blowing experience. (Spoiler alert: you don't have to channel your inner Will Smith while re-watching 'I, Robot' to prepare for the rise of the machines just yet!)
What's the 411 on Google I/O?
Google I/O (or simply I/O) is one of Google's huge developer conferences, which brings together developers from all over the globe and gives them a first look at Google's latest developer products and platforms. Basically, it's just a big festival for all things Google!
Interesting things learned
From a consumer-facing perspective, they showed off Google Duplex, whereby Google's AI voice-generated assistant makes phone calls on your behalf, realistically "faking" human interaction. I don't think it's fully baked yet, but certainly hints at where we're heading in the future. It definitely raised some interesting ethical questions about the nature of AI.
From a developer perspective, and one of the main reasons I went, I/O addressed a few interesting topics. These included the Polymer 3.0 launch, which represented simple but important steps forward for the Polymer Project (and the Web platform as a whole).
There were a whole bunch of talks on Progressive Web Apps (PWA), including new tooling, and interesting and revealing data on trends with PWA. Finally, there was a lot on machine learning!
Can you give a quick machine learning update?
Google announced ML Kit, which basically cuts across a whole bunch of its software offerings on Android and Web. This is to try and make it easier to access common machine learning data tasks, such as image and speech recognition, among other things. Essentially, it becomes a lot easier to do very sophisticated work without needing a PhD in machine learning.
As another example, Google showed off a JavaScript framework called Guess.js, which can use your Google Analytics data from your Web site to optimise the loading experience. A practical use case for machine learning to make your site faster for users.
All the talks were live-streamed and are available online. Which were some of your favourites?
I actually have three that I would recommend:
1. The state of the Web union. It was very insightful to see how things fit together on the Web.
2. The Polymer 3.0 roadmap. It talked about the new Polymer and where they are, together with discussing where they are with trying to drive the Web platform.
3. Ewa Gasperowicz and Addy Osmani talking about the basics of Web performance. They introduced the Guess.js framework.
What was your absolute favourite part of the entire event?
Easy answer. Everybody knows it is a massive festival with thousands of people everywhere. While it was easy to get caught up in bigger and busier areas, the most amazing parts were how they put so much attention into small details. There were graffiti walls with #IO18 logos, little Android robots driving around with selfie sticks, cotton candy machines in the corner of a random room and even Android sumo wrestling! There were domes showcasing things built with Google products, and my favourite was the Internet of things (IOT) one. It had robotic flowers that would follow your face as if your face was the sun, as well as flowers that changed colour based on your facial expressions, both using the 'Android Things' machine learning capabilities. So, for me, it was all about the little things that could so easily go unnoticed.
Any advice for future I/O attendees?
1. For my South African friends, nothing prepares you for the 27-hour flight and the jet-lag that follows!
2. When at Google's Mountain View headquarters, don't try and "borrow" a Google campus GBike to drive around. It's for Google employees only and you will be chased by security. But, this is, of course, what a little birdie told me; I'm innocent, I promise!
Share